One aspect of this that you may have already realized if you DIY your 3-part solutions is that a sodium bicarbonate (or sodium carbonate) solution when made up to the concentration typically used are very nearly saturated. That means that any evaporation at the tip of your dispenser is going to lead to crystallization.
There's several potential fixes. One is to make up the NaHCO3/Na2CO3 solution a bit more dilute than is typical, and simply up the dosing amount to compensate. Diluting it by 10% or so should be enough.
Another potential fix is to make sure that the cabinet underneath the tank stays closed to ensure high humidity, since reducing evaporation at the dispense tip will slow down crystallization.
Finally, since typical alk solutions are near saturation, and degree of saturation is a function of temperature (higher temps will dissolve more NaHCO3/Na2CO3), you could move your dosing container outside of the cabinet/sump. Presuming that your house temperature is a good bit lower than your reef tank's temperature, crystallization will occur in the dosing container rather than at the dispense tip.
One note about this. Depending on what sort of chemistry parameters you run, it might be possible that what you interpret as crystallization of sodium bicarbonate at the dispense tip isn't. Instead, it's actual formation of calcium carbonate from the reaction of the carbonate/bicarbonate with calcium in your tank water (even though the tip's above the water, there's still splashing/spray to consider). Fortunately, that's an easy test - scrape some of the crystals off of the line, and put them in tap water. If they dissolve, then it is indeed precipitation of the alkalinity reagent. OTOH, calcium carbonate will not dissolve in just about any amount of tap water. If you want further confirmation, simply add a few drops of acid to the test mixture. If the crystals won't dissolve in tap water, but quickly dissolve after a few drops of vinegar, then it's definitely calcium carbonate.
